| | 1 | @ vim:filetype=armasm\r |
| | 2 | @ some color conversion and blitting routines\r |
| | 3 | \r |
| | 4 | @ (c) Copyright 2006, 2007 notaz\r |
| | 5 | @ All Rights Reserved\r |
| | 6 | \r |
| | 7 | @ vim:filetype=armasm\r |
| | 8 | \r |
| | 9 | .text\r |
| | 10 | .align 4\r |
| | 11 | \r |
| | 12 | @ Convert 0000bbb0 ggg0rrr0 0000bbb0 ggg0rrr0\r |
| | 13 | @ to 00000000 rrr00000 ggg00000 bbb00000 ...\r |
| | 14 | \r |
| | 15 | @ lr = 0x00e000e0, out: r3=lower_pix, r2=higher_pix; trashes rin\r |
| | 16 | .macro convRGB32_2 rin sh=0\r |
| | 17 | and r2, lr, \rin, lsr #4 @ blue\r |
| | 18 | and r3, \rin, lr\r |
| | 19 | orr r2, r2, r3, lsl #8 @ g0b0g0b0\r |
| | 20 | \r |
| | 21 | mov r3, r2, lsl #16 @ g0b00000\r |
| | 22 | and \rin,lr, \rin, ror #12 @ 00r000r0 (reversed)\r |
| | 23 | orr r3, r3, \rin, lsr #16 @ g0b000r0\r |
| | 24 | .if \sh == 1\r |
| | 25 | mov r3, r3, ror #17 @ shadow mode\r |
| | 26 | .elseif \sh == 2\r |
| | 27 | adds r3, r3, #0x40000000 @ green\r |
| | 28 | orrcs r3, r3, #0xe0000000\r |
| | 29 | mov r3, r3, ror #8\r |
| | 30 | adds r3, r3, #0x40000000\r |
| | 31 | orrcs r3, r3, #0xe0000000\r |
| | 32 | mov r3, r3, ror #16\r |
| | 33 | adds r3, r3, #0x40000000\r |
| | 34 | orrcs r3, r3, #0xe0000000\r |
| | 35 | mov r3, r3, ror #24\r |
| | 36 | .else\r |
| | 37 | mov r3, r3, ror #16 @ r3=low\r |
| | 38 | .endif\r |
| | 39 | \r |
| | 40 | orr r3, r3, r3, lsr #3\r |
| | 41 | str r3, [r0], #4\r |
| | 42 | \r |
| | 43 | mov r2, r2, lsr #16\r |
| | 44 | orr r2, r2, \rin, lsl #16\r |
| | 45 | .if \sh == 1\r |
| | 46 | mov r2, r2, lsr #1\r |
| | 47 | .elseif \sh == 2\r |
| | 48 | mov r2, r2, ror #8\r |
| | 49 | adds r2, r2, #0x40000000 @ blue\r |
| | 50 | orrcs r2, r2, #0xe0000000\r |
| | 51 | mov r2, r2, ror #8\r |
| | 52 | adds r2, r2, #0x40000000\r |
| | 53 | orrcs r2, r2, #0xe0000000\r |
| | 54 | mov r2, r2, ror #8\r |
| | 55 | adds r2, r2, #0x40000000\r |
| | 56 | orrcs r2, r2, #0xe0000000\r |
| | 57 | mov r2, r2, ror #8\r |
| | 58 | .endif\r |
| | 59 | \r |
| | 60 | orr r2, r2, r2, lsr #3\r |
| | 61 | .if \sh == 1\r |
| | 62 | str r2, [r0, #0x40*2*4]\r |
| | 63 | .endif\r |
| | 64 | str r2, [r0], #4\r |
| | 65 | .endm\r |
| | 66 | \r |
| | 67 | \r |
| | 68 | .global bgr444_to_rgb32 @ void *to, void *from\r |
| | 69 | \r |
| | 70 | bgr444_to_rgb32:\r |
| | 71 | stmfd sp!, {r4-r7,lr}\r |
| | 72 | \r |
| | 73 | mov r12, #0x40>>3 @ repeats\r |
| | 74 | mov lr, #0x00e00000\r |
| | 75 | orr lr, lr, #0x00e0\r |
| | 76 | \r |
| | 77 | .loopRGB32:\r |
| | 78 | subs r12, r12, #1\r |
| | 79 | \r |
| | 80 | ldmia r1!, {r4-r7}\r |
| | 81 | convRGB32_2 r4\r |
| | 82 | convRGB32_2 r5\r |
| | 83 | convRGB32_2 r6\r |
| | 84 | convRGB32_2 r7\r |
| | 85 | bgt .loopRGB32\r |
| | 86 | \r |
| | 87 | ldmfd sp!, {r4-r7,pc}\r |
| | 88 | \r |
| | 89 | \r |
| | 90 | .global bgr444_to_rgb32_sh @ void *to, void *from\r |
| | 91 | \r |
| | 92 | bgr444_to_rgb32_sh:\r |
| | 93 | stmfd sp!, {r4-r7,lr}\r |
| | 94 | \r |
| | 95 | mov r12, #0x40>>3 @ repeats\r |
| | 96 | add r0, r0, #0x40*4\r |
| | 97 | mov lr, #0x00e00000\r |
| | 98 | orr lr, lr, #0x00e0\r |
| | 99 | \r |
| | 100 | .loopRGB32sh:\r |
| | 101 | subs r12, r12, #1\r |
| | 102 | \r |
| | 103 | ldmia r1!, {r4-r7}\r |
| | 104 | convRGB32_2 r4, 1\r |
| | 105 | convRGB32_2 r5, 1\r |
| | 106 | convRGB32_2 r6, 1\r |
| | 107 | convRGB32_2 r7, 1\r |
| | 108 | bgt .loopRGB32sh\r |
| | 109 | \r |
| | 110 | mov r12, #0x40>>3 @ repeats\r |
| | 111 | sub r1, r1, #0x40*2\r |
| | 112 | \r |
| | 113 | .loopRGB32hi:\r |
| | 114 | ldmia r1!, {r4-r7}\r |
| | 115 | convRGB32_2 r4, 2\r |
| | 116 | convRGB32_2 r5, 2\r |
| | 117 | convRGB32_2 r6, 2\r |
| | 118 | convRGB32_2 r7, 2\r |
| | 119 | \r |
| | 120 | subs r12, r12, #1\r |
| | 121 | bgt .loopRGB32hi\r |
| | 122 | \r |
| | 123 | ldmfd sp!, {r4-r7,lr}\r |
| | 124 | bx lr\r |
| | 125 | \r |
| | 126 | \r |
| | 127 | @ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@\r |
| | 128 | \r |
| | 129 | \r |
| | 130 | @ mode2 blitter\r |
| | 131 | .global vidcpy_m2 @ void *dest, void *src, int m32col, int with_32c_border\r |
| | 132 | vidcpy_m2:\r |
| | 133 | stmfd sp!, {r4-r6,lr}\r |
| | 134 | \r |
| | 135 | mov r12, #224 @ lines\r |
| | 136 | add r0, r0, #320*8\r |
| | 137 | add r1, r1, #8\r |
| | 138 | mov lr, #0\r |
| | 139 | \r |
| | 140 | tst r2, r2\r |
| | 141 | movne lr, #64\r |
| | 142 | tstne r3, r3\r |
| | 143 | addne r0, r0, #32\r |
| | 144 | \r |
| | 145 | vidCpyM2_loop_out:\r |
| | 146 | mov r6, #10\r |
| | 147 | sub r6, r6, lr, lsr #5 @ -= 2 in 32col mode\r |
| | 148 | vidCpyM2_loop:\r |
| | 149 | subs r6, r6, #1\r |
| | 150 | ldmia r1!, {r2-r5}\r |
| | 151 | stmia r0!, {r2-r5}\r |
| | 152 | ldmia r1!, {r2-r5}\r |
| | 153 | stmia r0!, {r2-r5}\r |
| | 154 | bne vidCpyM2_loop\r |
| | 155 | \r |
| | 156 | subs r12,r12,#1\r |
| | 157 | add r0, r0, lr\r |
| | 158 | add r1, r1, #8\r |
| | 159 | add r1, r1, lr\r |
| | 160 | bne vidCpyM2_loop_out\r |
| | 161 | \r |
| | 162 | ldmfd sp!, {r4-r6,pc}\r |
| | 163 | \r |
| | 164 | \r |
| | 165 | .global vidcpy_m2_rot @ void *dest, void *src, int m32col, int with_32c_border\r |
| | 166 | vidcpy_m2_rot:\r |
| | 167 | stmfd sp!,{r4-r8,lr}\r |
| | 168 | add r1, r1, #8\r |
| | 169 | tst r2, r2\r |
| | 170 | subne r1, r1, #32 @ adjust\r |
| | 171 | \r |
| | 172 | mov r4, r0\r |
| | 173 | mov r5, r1\r |
| | 174 | mov r6, r2\r |
| | 175 | mov r7, #8+4\r |
| | 176 | \r |
| | 177 | vidcpy_m2_rot_loop:\r |
| | 178 | @ a bit lame but oh well..\r |
| | 179 | mov r0, r4\r |
| | 180 | mov r1, r5\r |
| | 181 | mov r2, r7\r |
| | 182 | mov r3, r6\r |
| | 183 | mov r8, #328\r |
| | 184 | adr lr, after_rot_blit8\r |
| | 185 | stmfd sp!,{r4-r8,lr}\r |
| | 186 | b rotated_blit8_2\r |
| | 187 | \r |
| | 188 | after_rot_blit8:\r |
| | 189 | add r5, r5, #328*4\r |
| | 190 | add r7, r7, #4\r |
| | 191 | cmp r7, #224+8+4\r |
| | 192 | ldmgefd sp!,{r4-r8,pc}\r |
| | 193 | b vidcpy_m2_rot_loop\r |
| | 194 | \r |
| | 195 | \r |
| | 196 | .global rotated_blit8 @ void *dst, void *linesx4, u32 y, int is_32col\r |
| | 197 | rotated_blit8:\r |
| | 198 | stmfd sp!,{r4-r8,lr}\r |
| | 199 | mov r8, #320\r |
| | 200 | \r |
| | 201 | rotated_blit8_2:\r |
| | 202 | add r0, r0, #(240*320)\r |
| | 203 | sub r0, r0, #(240+4) @ y starts from 4\r |
| | 204 | add r0, r0, r2\r |
| | 205 | \r |
| | 206 | tst r3, r3\r |
| | 207 | subne r0, r0, #(240*32)\r |
| | 208 | addne r1, r1, #32\r |
| | 209 | movne lr, #256/4\r |
| | 210 | moveq lr, #320/4\r |
| | 211 | \r |
| | 212 | rotated_blit_loop8:\r |
| | 213 | mov r6, r1\r |
| | 214 | ldr r2, [r6], r8\r |
| | 215 | ldr r3, [r6], r8\r |
| | 216 | ldr r4, [r6], r8\r |
| | 217 | ldr r5, [r6], r8\r |
| | 218 | \r |
| | 219 | mov r6, r2, lsl #24\r |
| | 220 | mov r6, r6, lsr #8\r |
| | 221 | orr r6, r6, r3, lsl #24\r |
| | 222 | mov r6, r6, lsr #8\r |
| | 223 | orr r6, r6, r4, lsl #24\r |
| | 224 | mov r6, r6, lsr #8\r |
| | 225 | orr r6, r6, r5, lsl #24\r |
| | 226 | str r6, [r0], #-240\r |
| | 227 | \r |
| | 228 | and r6, r3, #0xff00\r |
| | 229 | and r7, r2, #0xff00\r |
| | 230 | orr r6, r6, r7, lsr #8\r |
| | 231 | and r7, r4, #0xff00\r |
| | 232 | orr r6, r6, r7, lsl #8\r |
| | 233 | and r7, r5, #0xff00\r |
| | 234 | orr r6, r6, r7, lsl #16\r |
| | 235 | str r6, [r0], #-240\r |
| | 236 | \r |
| | 237 | and r6, r4, #0xff0000\r |
| | 238 | and r7, r2, #0xff0000\r |
| | 239 | orr r6, r6, r7, lsr #16\r |
| | 240 | and r7, r3, #0xff0000\r |
| | 241 | orr r6, r6, r7, lsr #8\r |
| | 242 | and r7, r5, #0xff0000\r |
| | 243 | orr r6, r6, r7, lsl #8\r |
| | 244 | str r6, [r0], #-240\r |
| | 245 | \r |
| | 246 | mov r6, r5, lsr #24\r |
| | 247 | mov r6, r6, lsl #8\r |
| | 248 | orr r6, r6, r4, lsr #24\r |
| | 249 | mov r6, r6, lsl #8\r |
| | 250 | orr r6, r6, r3, lsr #24\r |
| | 251 | mov r6, r6, lsl #8\r |
| | 252 | orr r6, r6, r2, lsr #24\r |
| | 253 | str r6, [r0], #-240\r |
| | 254 | \r |
| | 255 | subs lr, lr, #1\r |
| | 256 | add r1, r1, #4\r |
| | 257 | bne rotated_blit_loop8\r |
| | 258 | \r |
| | 259 | ldmfd sp!,{r4-r8,pc}\r |
| | 260 | \r |
| | 261 | \r |
| | 262 | @ input: r2-r5\r |
| | 263 | @ output: r7,r8\r |
| | 264 | @ trash: r6\r |
| | 265 | .macro rb_line_low\r |
| | 266 | mov r6, r2, lsl #16\r |
| | 267 | mov r7, r3, lsl #16\r |
| | 268 | orr r7, r7, r6, lsr #16\r |
| | 269 | mov r6, r4, lsl #16\r |
| | 270 | mov r8, r5, lsl #16\r |
| | 271 | orr r8, r8, r6, lsr #16\r |
| | 272 | .endm\r |
| | 273 | \r |
| | 274 | .macro rb_line_hi\r |
| | 275 | mov r6, r2, lsr #16\r |
| | 276 | mov r7, r3, lsr #16\r |
| | 277 | orr r7, r6, r7, lsl #16\r |
| | 278 | mov r6, r4, lsr #16\r |
| | 279 | mov r8, r5, lsr #16\r |
| | 280 | orr r8, r6, r8, lsl #16\r |
| | 281 | .endm\r |
| | 282 | \r |
| | 283 | .global rotated_blit16 @ void *dst, void *linesx4, u32 y, int is_32col\r |
| | 284 | rotated_blit16:\r |
| | 285 | stmfd sp!,{r4-r8,lr}\r |
| | 286 | \r |
| | 287 | add r0, r0, #(240*320)*2\r |
| | 288 | sub r0, r0, #(240+4)*2 @ y starts from 4\r |
| | 289 | add r0, r0, r2, lsl #1\r |
| | 290 | \r |
| | 291 | tst r3, r3\r |
| | 292 | subne r0, r0, #(240*32)*2\r |
| | 293 | addne r1, r1, #32*2\r |
| | 294 | movne lr, #256/4\r |
| | 295 | moveq lr, #320/4\r |
| | 296 | \r |
| | 297 | rotated_blit_loop16:\r |
| | 298 | ldr r2, [r1, #320*0*2]\r |
| | 299 | ldr r3, [r1, #320*1*2]\r |
| | 300 | ldr r4, [r1, #320*2*2]\r |
| | 301 | ldr r5, [r1, #320*3*2]\r |
| | 302 | rb_line_low\r |
| | 303 | stmia r0, {r7,r8}\r |
| | 304 | sub r0, r0, #240*2\r |
| | 305 | rb_line_hi\r |
| | 306 | stmia r0, {r7,r8}\r |
| | 307 | sub r0, r0, #240*2\r |
| | 308 | \r |
| | 309 | ldr r2, [r1, #320*0*2+4]\r |
| | 310 | ldr r3, [r1, #320*1*2+4]\r |
| | 311 | ldr r4, [r1, #320*2*2+4]\r |
| | 312 | ldr r5, [r1, #320*3*2+4]\r |
| | 313 | rb_line_low\r |
| | 314 | stmia r0, {r7,r8}\r |
| | 315 | sub r0, r0, #240*2\r |
| | 316 | rb_line_hi\r |
| | 317 | stmia r0, {r7,r8}\r |
| | 318 | sub r0, r0, #240*2\r |
| | 319 | \r |
| | 320 | subs lr, lr, #1\r |
| | 321 | add r1, r1, #8\r |
| | 322 | bne rotated_blit_loop16\r |
| | 323 | \r |
| | 324 | ldmfd sp!,{r4-r8,pc}\r |
| | 325 | \r |
| | 326 | \r |
| | 327 | .global spend_cycles @ c\r |
| | 328 | \r |
| | 329 | spend_cycles:\r |
| | 330 | mov r0, r0, lsr #2 @ 4 cycles/iteration\r |
| | 331 | sub r0, r0, #2 @ entry/exit/init\r |
| | 332 | .sc_loop:\r |
| | 333 | subs r0, r0, #1\r |
| | 334 | bpl .sc_loop\r |
| | 335 | \r |
| | 336 | bx lr\r |
| | 337 | \r |
notaz.gp2x.de / picodrive.git / blame_incremental